Abstract
Background
Delay in access to primary total hip (THA) arthroplasty continues to pose a substantial burden to patients and society in publicly funded healthcare systems. The majority of strategies to decrease wait times have focused on the time from surgical consult to surgery, however a large proportion of total wait time for these patients is the time from primary care referral to surgical consultation. Prioritization scoring tools and patient reported outcome measures are being used in an attempt to ration limited resources in the face of increasing demand. However, to our knowledge, no study has investigated whether a referral Oxford Hip Score (OHS) could be used to triage non-surgical referrals appropriately, in an effort to increase timely access to specialists for patients that are candidates for total joint replacement (TJR).
Purpose
1) To determine if a referral OHS has the predictive ability to discriminate when a hip patient will be deemed surgical versus conservative by the surgeon during their first consultation 2) To identify an OHS cut-off point that can be used to accurately predict when a primary THA referral will be deemed conservative by the consultant surgeon during the first consultation.
Methods
We retrospectively reviewed all consecutive THA consultations from a single surgeon's tertiary, high volume practice over a 3-year period. Patients with a pre-consultation OHS, BMI <41, and no absolute contraindication to TJR were included. Consultation were categorized into two groups based on surgeon's decision, those that were offered THR during their first consultation (operative) versus those that were not (conservative). Baseline demographic data and OHS were abstracted. Variables of interest were compared between cohorts using the exact chi-square test and Wilcoxon rank-sum test. Spearman's rank correlation coefficients were used to measure association between pre-consult OHS and the surgeon's decision. A receiver operator characteristic (ROC) curve analysis was used to calculate the area under the curve (AUC) and to identify a cut-off point for the pre-operative OHS that identified whether or not a referral was deemed conservative.
Results
The study 478 hips (388 patients) with a median OHS of 22 (IQR 16–29). Median pre-consultation OHS demonstrated clinically and statistically significant differences between the surgical versus conservative cohorts (p<0.001). Spearman's rank correlation coefficient between OHS and a patient being deemed surgical or conservative was strong for the OHS at −0.62 (95% CI −0.67 to −0.56). The ROC AUC values for hip consults (0.87, 95% CI 0.84–0.91) was good, indicating that pre-consult OHS has predictive ability to discriminate a surgeon's decision of surgical versus conservative. One plausible conservative threshold that optimized sensitivity and NPV for hips is OHS >34 (sensitivity=0.997 NPV=0.978). ROC analysis identified severable potential lower, depending on weight of prioritization of sensitivity, specificity, and NPV.
Conclusion
Referral OHS demonstrate good ability to discriminate when a knee or hip TJR referral will be deemed non-surgical versus surgical at their first consultation in a single surgeon's practice. Multiple potential OHS thresholds can be applied as a tool to decrease wait times for primary THR. However, a cost analysis would aid in identifying the optimal cut-off score, and these findings need to be validated with multi-surgeon/center studies before they can be broadly applied.
